3
SLj~o @ s d Z dZddlZddlZddlmZ dZe Zdd Zdd Z d
d Z
d d
Z dd Z dd Z
e e Z ZZe e ZZe e ej ZZe
d Z ZZe
d Z ZZe
dZ dd Z dd Z dd Z dd Z d d Z!d d Z"d!d" Z#d#d$ Z$e# Z%e$ Z&d%d& Z'd'd( Z(d)d* Z)d+d, Z*d-d. Z+d/d0 Z,e)ej-e&e Z. Z/Z0e)ej-e%e Z1Z2e*ej-e%eej Z3Z4e+ej5d1Z6e+ej7d2Z8e+ej5d3Z9e+ej7d4Z:e,ej5d5Z;e,ej7d6Zd;d< Z?d=d> Z@d?d@ ZAdAdB ZBdCdD ZCdS )Ea Code for encoding protocol message primitives.
Contains the logic for encoding every logical protocol field type
into one of the 5 physical wire types.
This code is designed to push the Python interpreter's performance to the
limits.
The basic idea is that at startup time, for every field (i.e. every
FieldDescriptor) we construct two functions: a "sizer" and an "encoder". The
sizer takes a value of this field's type and computes its byte size. The
encoder takes a writer function and a value. It encodes the value into byte
strings and invokes the writer function to write those strings. Typically the
writer function is the write() method of a BytesIO.
We try to do as much work as possible when constructing the writer and the
sizer rather than when calling them. In particular:
* We copy any needed global functions to local variables, so that we do not need
to do costly global table lookups at runtime.
* Similarly, we try to do any attribute lookups at startup time if possible.
* Every field's tag is encoded to bytes at startup, since it can't change at
runtime.
* Whatever component of the field size we can compute at startup, we do.
* We *avoid* sharing code if doing so would make the code slower and not sharing
does not burden us too much. For example, encoders for repeated fields do
not just call the encoders for singular fields in a loop because this would
add an extra function call overhead for every loop iteration; instead, we
manually inline the single-value encoder into the loop.
* If a Python function lacks a return statement, Python actually generates
instructions to pop the result of the last statement off the stack, push
None onto the stack, and then return that. If we really don't care what
value is returned, then we can save two instructions by returning the
result of the last statement. It looks funny but it helps.
* We assume that type and bounds checking has happened at a higher level.
z kenton@google.com (Kenton Varda) N) wire_formatg c C sp | dkr dS | dkrdS | dkr$dS | dkr0dS | d kr.SpecificSizer..PackedFieldSizec s* t | }x| D ]}| |7 }qW |S )N)len)r r r )r r r r RepeatedFieldSize s
z>_SimpleSizer..SpecificSizer..RepeatedFieldSizec s | S )Nr )r )r r r r FieldSize s z6_SimpleSizer..SpecificSizer..FieldSize)r r )r is_repeated is_packedr r r )r )r r r
SpecificSizer s z#_SimpleSizer..SpecificSizerr )r r! r )r r _SimpleSizer~ s r" c s fdd}|S )zLike SimpleSizer, but modify_value is invoked on each value before it is
passed to compute_value_size. modify_value is typically ZigZagEncode.c sV t | |r&t fdd}|S |r>fdd}|S fdd}|S d S )Nc s2 d}x | D ]}| |7 }q
W || S )Nr r )r r r )r r modify_valuer r r r s
z>_ModifiedSizer..SpecificSizer..PackedFieldSizec s. t | }x | D ]}| |7 }qW |S )N)r )r r r )r r# r r r r s
z@_ModifiedSizer..SpecificSizer..RepeatedFieldSizec s | S )Nr )r )r r# r r r r s z8_ModifiedSizer..SpecificSizer..FieldSize)r r )r r r r r r )r r# )r r r r! s z%_ModifiedSizer..SpecificSizerr )r r# r! r )r r# r _ModifiedSizer s r$ c s fdd}|S )zWLike _SimpleSizer except for a fixed-size field. The input is the size
of one value.c s\ t | |r$tfdd}|S |r@ fdd}|S fdd}|S d S )Nc s t | }| | S )N)r )r r )r r
value_sizer r r s z;_FixedSizer..SpecificSizer..PackedFieldSizec s t | S )N)r )r ) element_sizer r r s z=_FixedSizer..SpecificSizer..RepeatedFieldSizec s S )Nr )r )
field_sizer r r s z5_FixedSizer..SpecificSizer..FieldSize)r r )r r r r r r )r% )r& r' r r r r! s z"_FixedSizer..SpecificSizerr )r% r! r )r% r _FixedSizer s r( r r r c s@ t | t t|r( fdd}|S fdd}|S dS )z#Returns a sizer for a string field.c s< t | }x*| D ]"}|jd}| || 7 }qW |S )Nzutf-8)r encode)r r r l)r local_lenr r r r s
z&StringSizer..RepeatedFieldSizec s | j d} | | S )Nzutf-8)r) )r r* )r r+ r r r r s z StringSizer..FieldSizeN)r r r )r r r r r r )r r+ r r StringSizer s r, c s@ t | t t|r( fdd}|S fdd}|S dS )z"Returns a sizer for a bytes field.c s6 t | }x$| D ] }|}| || 7 }qW |S )N)r )r r r r* )r r+ r r r r s
z%BytesSizer..RepeatedFieldSizec s | } | | S )Nr )r r* )r r+ r r r r s z BytesSizer..FieldSizeN)r r r )r r r r r r )r r+ r r
BytesSizer s r- c s4 t | d |r fdd}|S fdd}|S dS )z"Returns a sizer for a group field.r c s* t | }x| D ]}||j 7 }qW |S )N)r ByteSize)r r r )r r r r s
z%GroupSizer..RepeatedFieldSizec s | j S )N)r. )r )r r r r s z GroupSizer..FieldSizeN)r )r r r r r r )r r
GroupSizer s r/ c s8 t | t |r" fdd}|S fdd}|S dS )z$Returns a sizer for a message field.c s6 t | }x$| D ] }|j }| || 7 }qW |S )N)r r. )r r r r* )r r r r r + s
z'MessageSizer..RepeatedFieldSizec s | j } | | S )N)r. )r r* )r r r r r 3 s z MessageSizer..FieldSizeN)r r )r r r r r r )r r r MessageSizer$ s r0 c s: t dd t d t| t d t fdd}|S )zReturns a sizer for extensions of MessageSet.
The message set message looks like this:
message MessageSet {
repeated group Item = 1 {
required int32 type_id = 2;
required string message = 3;
}
}
r r r c s | j } | | S )N)r. )r r* )r static_sizer r r L s z&MessageSetItemSizer..FieldSize)r r )r r r )r r1 r MessageSetItemSizer= s
r2 c s( | j t| jdd fdd}|S )z Returns a sizer for a map field.Fc sD d}x:| D ]2}| | }j ||d}||7 } r
|j q
W |S )Nr )keyr )_concrete_classr. )Z map_valueZtotalr3 r entry_msg)is_message_map
message_sizer message_typer r r _ s
zMapSizer..FieldSize)r8 r0 number)field_descriptorr6 r r )r6 r7 r8 r MapSizerW s r; c C s dd } | S )zBReturn an encoder for a basic varint value (does not include tag).c S sJ |d@ }|dL }x*|r:| t jd|B |d@ }|dL }qW | t j|S )Nr r
)sixint2byte)writer unused_deterministicbitsr r r EncodeVarintw s z$_VarintEncoder..EncodeVarintr )rB r r r _VarintEncodert s rC c C s dd } | S )zKReturn an encoder for a basic signed varint value (does not include
tag).c S sZ |dk r|d7 }|d@ }|dL }x*|rJ| t jd|B |d@ }|dL }q"W | t j|S )Nr r @ r r
r< l )r= r> )r? r r@ rA r r r EncodeSignedVarint s z0_SignedVarintEncoder..EncodeSignedVarintr )rE r r r _SignedVarintEncoder s rF c C s g }t |j| d dj|S )zEncode the given integer as a varint and return the bytes. This is only
called at startup time so it doesn't need to be fast.T )
_EncodeVarintappendjoin)r piecesr r r _VarintBytes s rL c C s t jttj| |S )zCEncode the given tag and return the bytes. Only called at startup.)r= Z binary_typerL r r )r wire_typer r r TagBytes s rN c s fdd}|S )a_ Return a constructor for an encoder for fields of a particular type.
Args:
wire_type: The field's wire type, for encoding tags.
encode_value: A function which encodes an individual value, e.g.
_EncodeVarint().
compute_value_size: A function which computes the size of an individual
value, e.g. _VarintSize().
c sj |r*t | tjt fdd}|S |rJt | fdd}|S t | fdd}|S d S )Nc sP | d}x|D ]}| |7 }qW | || x|D ]}| || q8W d S )Nr r )r? r
deterministicsizer )r encode_valuelocal_EncodeVarint tag_bytesr r EncodePackedField s
zB_SimpleEncoder..SpecificEncoder..EncodePackedFieldc s& x |D ]}| | || qW d S )Nr )r? r rO r )rQ rS r r EncodeRepeatedField s
zD_SimpleEncoder..SpecificEncoder..EncodeRepeatedFieldc s | | ||S )Nr )r? r rO )rQ rS r r EncodeField s z<_SimpleEncoder..SpecificEncoder..EncodeField)rN r WIRETYPE_LENGTH_DELIMITEDrH )r r r rT rU rV )r rQ rM )rR rS r SpecificEncoder s
z'_SimpleEncoder..SpecificEncoderr )rM rQ r rX r )r rQ rM r _SimpleEncoder s rY c s fdd}|S )zLike SimpleEncoder but additionally invokes modify_value on every value
before passing it to encode_value. Usually modify_value is ZigZagEncode.c sp |r,t | tjt fdd}|S |rNt | fdd}|S t | fdd}|S d S )Nc sX | d}x |D ]}| |7 }qW | || x |D ]}| || q.SpecificEncoder..EncodePackedFieldc s* x$|D ] }| | || qW d S )Nr )r? r rO r )rQ r# rS r r rU s
zF_ModifiedEncoder..SpecificEncoder..EncodeRepeatedFieldc s | | ||S )Nr )r? r rO )rQ r# rS r r rV s z>_ModifiedEncoder..SpecificEncoder..EncodeField)rN r rW rH )r r r rT rU rV )r rQ r# rM )rR rS r rX s
z)_ModifiedEncoder..SpecificEncoderr )rM rQ r r# rX r )r rQ r# rM r _ModifiedEncoder s rZ c s t j fdd}|S )zReturn a constructor for an encoder for a fixed-width field.
Args:
wire_type: The field's wire type, for encoding tags.
format: The format string to pass to struct.pack().
c sv t j|r2t| tjt fdd}|S |rTt| fdd}|S t| fdd}|S d S )Nc s< | | t | | x|D ]}| | q"W d S )N)r )r? r rO r )formatrR local_struct_packrS r% r r rT s
zF_StructPackEncoder..SpecificEncoder..EncodePackedFieldc s( x"|D ]}| | | qW d S )Nr )r? r r@ r )r[ r\ rS r r rU s
zH_StructPackEncoder..SpecificEncoder..EncodeRepeatedFieldc s | | |S )Nr )r? r r@ )r[ r\ rS r r rV s z@_StructPackEncoder..SpecificEncoder..EncodeField)structpackrN r rW rH )r r r rT rU rV )r[ r% rM )rR r\ rS r rX s
z+_StructPackEncoder..SpecificEncoder)r] calcsize)rM r[ rX r )r[ r% rM r _StructPackEncoder s
r` c sP t jdkr dd n dkr.dd n td fdd}|S ) aq Return a constructor for an encoder for float fields.
This is like StructPackEncoder, but catches errors that may be due to
passing non-finite floating-point values to struct.pack, and makes a
second attempt to encode those values.
Args:
wire_type: The field's wire type, for encoding tags.
format: The format string to pass to struct.pack().
r c S s< |t kr| d n&|tkr$| d n||kr6| d n d S )Ns s s )_POS_INF_NEG_INF)r? r r r r EncodeNonFiniteOrRaise+ s
z5_FloatingPointEncoder..EncodeNonFiniteOrRaiser c S s< |t kr| d n&|tkr$| d n||kr6| d n d S )Ns s s )ra rb )r? r r r r rc 6 s
zGCan't encode floating-point values that are %d bytes long (only 4 or 8)c s| t j|r4t| tjt fdd}|S |rXt| fdd}|S t| fdd}|S d S )Nc s` | | t | | x>|D ]6}y| | W q" tk
rV | | Y q"X q"W d S )N)r SystemError)r? r rO r )rc r[ rR r\ rS r% r r rT H s
zI_FloatingPointEncoder..SpecificEncoder..EncodePackedFieldc sL xF|D ]>}| y| | W q t k
rB | | Y qX qW d S )N)rd )r? r r@ r )rc r[ r\ rS r r rU U s
zK_FloatingPointEncoder..SpecificEncoder..EncodeRepeatedFieldc s> | y| | W n t k
r8 | | Y nX d S )N)rd )r? r r@ )rc r[ r\ rS r r rV _ s
zC_FloatingPointEncoder..SpecificEncoder..EncodeField)r] r^ rN r rW rH )r r r rT rU rV )rc r[ r% rM )rR r\ rS r rX C s
z._FloatingPointEncoder..SpecificEncoder)r] r_
ValueError)rM r[ rX r )rc r[ r% rM r _FloatingPointEncoder s
$rf z.EncodePackedFieldc s0 x*|D ]"}| |r | q| qW d S )Nr )r? r r@ r )ri rS rj r r rU s
z(BoolEncoder..EncodeRepeatedFieldc s | |r| S | S )Nr )r? r r@ )ri rS rj r r rV s z BoolEncoder..EncodeFieldN)rN r rW rH WIRETYPE_VARINT)r r r rT rU rV r )ri rR rS rj r BoolEncoder s rl c sD t | tjt t|r, fdd}|S fdd}|S dS )z&Returns an encoder for a string field.c s< x6|D ].}|j d}| | || | | qW d S )Nzutf-8)r) )r? r rO r encoded)rR r+ tagr r rU s
z*StringEncoder..EncodeRepeatedFieldc s* |j d}| | || | |S )Nzutf-8)r) )r? r rO rm )rR r+ rn r r rV s
z"StringEncoder..EncodeFieldN)rN r rW rH r )r r r rU rV r )rR r+ rn r
StringEncoder s ro c sD t | tjt t|r, fdd}|S fdd}|S dS )z%Returns an encoder for a bytes field.c s2 x,|D ]$}| | || | | qW d S )Nr )r? r rO r )rR r+ rn r r rU s
z)BytesEncoder..EncodeRepeatedFieldc s | | || | |S )Nr )r? r rO )rR r+ rn r r rV s z!BytesEncoder..EncodeFieldN)rN r rW rH r )r r r rU rV r )rR r+ rn r BytesEncoder s rp c sD t | tjt | tj |r. fdd}|S fdd}|S dS )z%Returns an encoder for a group field.c s. x(|D ] }| |j | | | qW d S )N)_InternalSerialize)r? r rO r )end_tag start_tagr r rU s
z)GroupEncoder..EncodeRepeatedFieldc s | |j | | | S )N)rq )r? r rO )rr rs r r rV s z!GroupEncoder..EncodeFieldN)rN r WIRETYPE_START_GROUPWIRETYPE_END_GROUP)r r r rU rV r )rr rs r GroupEncoder s rv c s< t | tjt |r& fdd}|S fdd}|S dS )z'Returns an encoder for a message field.c s6 x0|D ](}| | |j | |j| | qW d S )N)r. rq )r? r rO r )rR rn r r rU s
z+MessageEncoder..EncodeRepeatedFieldc s$ | | |j | |j| |S )N)r. rq )r? r rO )rR rn r r rV s z#MessageEncoder..EncodeFieldN)rN r rW rH )r r r rU rV r )rR rn r MessageEncoder s rw c sR dj tdtjtdtjt| tdtjgtdtj t fdd}|S )zEncoder for extensions of MessageSet.
The message set message looks like this:
message MessageSet {
repeated group Item = 1 {
required int32 type_id = 2;
required string message = 3;
}
}
rG r r r c s, | | |j | |j| | | S )N)r. rq )r? r rO ) end_bytesrR start_bytesr r rV s z*MessageSetItemEncoder..EncodeField) rJ rN r rt rk rL rW ru rH )r rV r )rx rR ry r MessageSetItemEncoder s
rz c s& | j t| jdd fdd}|S )zEncoder for extensions of MessageSet.
Maps always have a wire format like this:
message MapEntry {
key_type key = 1;
value_type value = 2;
}
repeated MapEntry map = N;
Fc sD |rt |j n|}x*|D ]"}j||| d} | || qW d S )N)r3 r )sortedkeysr4 )r? r rO Z
value_keysr3 r5 )encode_messager8 r r rV 5 s
z MapEncoder..EncodeField)r8 rw r9 )r: rV r )r} r8 r
MapEncoder& s r~ )D__doc__
__author__r] r= Zgoogle.protobuf.internalr ra rb r r r r" r$ r( Z
Int32SizerZ
Int64SizerZ EnumSizerZ UInt32SizerZ UInt64SizerZ ZigZagEncodeZ SInt32SizerZ SInt64SizerZ Fixed32SizerZ
SFixed32SizerZ
FloatSizerZ Fixed64SizerZ
SFixed64SizerZ DoubleSizerZ BoolSizerr, r- r/ r0 r2 r; rC rF rH Z_EncodeSignedVarintrL rN rY rZ r` rf rk Z Int32EncoderZ Int64EncoderZ EnumEncoderZ
UInt32EncoderZ
UInt64EncoderZ
SInt32EncoderZ
SInt64EncoderZWIRETYPE_FIXED32ZFixed32EncoderZWIRETYPE_FIXED64ZFixed64EncoderZSFixed32EncoderZSFixed64EncoderZ FloatEncoderZ
DoubleEncoderrl ro rp rv rw rz r~ r r r r A sn !
)"&R %